You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@beam.apache.org by Chuong Nguyen <ch...@acvconsultancy.com> on 2020/08/04 07:15:23 UTC

Could not run side_input in a streaming pipeline

Hi dev team,
I want to add a side_input into a ParDo transform. Side_input is a table
from Bigquery.
However, I am facing a weird issue. I could run this pipeline with
*DirectRunner* on a local machine but I am not able to run this pipeline
with *DataflowRunner. *The error message is:
```
Job did not reach to a terminal state after waiting indefinitely
```
And it traced into my *sales_transaction *table.
Attachment is the pipeline issue.

I am looking for help from the community.
Thanks and best regards.

Re: Could not run side_input in a streaming pipeline

Posted by Pablo Estrada <pa...@google.com>.
Hi Chuong!
Are you able to share more information about what Dataflow is telling you
in the UI? (Are steps running? Are there any elements being processed?)
How big is your BQ table?
Best
-P.

On Tue, Aug 4, 2020 at 9:39 AM Chuong Nguyen <
chuong.nguyen@acvconsultancy.com> wrote:

> Hi dev team,
> I want to add a side_input into a ParDo transform. Side_input is a table
> from Bigquery.
> However, I am facing a weird issue. I could run this pipeline with
> *DirectRunner* on a local machine but I am not able to run this pipeline
> with *DataflowRunner. *The error message is:
> ```
> Job did not reach to a terminal state after waiting indefinitely
> ```
> And it traced into my *sales_transaction *table.
> Attachment is the pipeline issue.
>
> I am looking for help from the community.
> Thanks and best regards.
>